Quantum annealing and the variational quantum eigensolver are two promising quantum algorithms to find the ground state of complicated Hamiltonians on near-term quantum devices. However, it is necessary to limit the evolution time or the circuit depth as much as possible since otherwise decoherence will degrade the computation. Even when this is done, there always exists a non-negligible estimation error in the ground state energy. Here we propose a scalable extrapolation approach to mitigate this error. With an appropriate regression, we can significantly improve the estimation accuracy for quantum annealing and variational quantum eigensolver for fixed quantum resources. The inference is achieved by extrapolating the annealing time to infinity or extrapolating the variance to zero. The only additional overhead is an increase in the number of measurements by a constant factor. We verified the validity of our method with the transverse-field Ising model. The method is robust to noise, and the techniques are applicable to other physics problems. Analytic derivations for the quadratic convergence feature of the residual energy in quantum annealing and the linear convergence feature of energy variance are given.
The quantum approximate optimization algorithm (QAOA) transforms a simple many-qubit wavefunction into one which encodes the solution to a difficult classical optimization problem. It does this by optimizing the schedule according to which two unitary operators are alternately applied to the qubits. In this paper, this procedure is modified by updating the operators themselves to include local fields, using information from the measured wavefunction at the end of one iteration step to improve the operators at later steps. It is shown by numerical simulation on MAXCUT problems that this decreases the runtime of QAOA very substantially. This improvement appears to increase with the problem size. Our method requires essentially the same number of quantum gates per optimization step as the standard QAOA. Application of this modified algorithm should bring closer the time to quantum advantage for optimization problems.
Zero-Shot Classification (ZSC) equips the learned model with the ability to recognize the visual instances from the novel classes via constructing the interactions between the visual and the semantic modalities. In contrast to the traditional image classification, ZSC is easily suffered from the class-imbalance issue since it is more concerned with the class-level knowledge transfer capability. In the real world, the class samples follow a long-tailed distribution, and the discriminative information in the sample-scarce seen classes is hard to be transferred to the related unseen classes in the traditional batch-based training manner, which degrades the overall generalization ability a lot. Towards alleviating the class imbalance issue in ZSC, we propose a sample-balanced training process to encourage all training classes to contribute equally to the learned model. Specifically, we randomly select the same number of images from each class across all training classes to form a training batch to ensure that the sample-scarce classes contribute equally as those classes with sufficient samples during each iteration. Considering that the instances from the same class differ in class representativeness, we further develop an efficient semantics-guided feature fusion model to obtain discriminative class visual prototype for the following visual-semantic interaction process via distributing different weights to the selected samples based on their class representativeness. Extensive experiments on three imbalanced ZSC benchmark datasets for both the Traditional ZSC (TZSC) and the Generalized ZSC (GZSC) tasks demonstrate our approach achieves promising results especially for the unseen categories those are closely related to the sample-scarce seen categories.
Existing generative Zero-Shot Learning (ZSL) methods only consider the unidirectional alignment from the class semantics to the visual features while ignoring the alignment from the visual features to the class semantics, which fails to construct the visual-semantic interactions well. In this paper, we propose to synthesize visual features based on an auto-encoder framework paired with bi-adversarial networks respectively for visual and semantic modalities to reinforce the visual-semantic interactions with a bi-directional alignment, which ensures the synthesized visual features to fit the real visual distribution and to be highly related to the semantics. The encoder aims at synthesizing real-like visual features while the decoder forces both the real and the synthesized visual features to be more related to the class semantics. To further capture the discriminative information of the synthesized visual features, both the real and synthesized visual features are forced to be classified into the correct classes via a classification network. Experimental results on four benchmark datasets show that the proposed approach is particularly competitive on both the traditional ZSL and the generalized ZSL tasks.
As an important and challenging problem in computer vision, zero-shot learning (ZSL) aims at automatically recognizing the instances from unseen object classes without training data. To address this problem, ZSL is usually carried out in the following two aspects: 1) capturing the domain distribution connections between seen classes data and unseen classes data; and 2) modeling the semantic interactions between the image feature space and the label embedding space. Motivated by these observations, we propose a bidirectional mapping based semantic relationship modeling scheme that seeks for crossmodal knowledge transfer by simultaneously projecting the image features and label embeddings into a common latent space. Namely, we have a bidirectional connection relationship that takes place from the image feature space to the latent space as well as from the label embedding space to the latent space. To deal with the domain shift problem, we further present a transductive learning approach that formulates the class prediction problem in an iterative refining process, where the object classification capacity is progressively reinforced through bootstrapping-based model updating over highly reliable instances. Experimental results on three benchmark datasets (AwA, CUB and SUN) demonstrate the effectiveness of the proposed approach against the state-of-the-art approaches.
